RFR: JDK-8302043: [ AIX ] Safefetch fails for bad_addressN and bad_address32

[Varada M]

On Wed, 8 Feb 2023 09:52:36 GMT, Thomas Stuefe <stuefe at openjdk.org> wrote:

> Hi @varada1110,
> 
> thank you. But could you please adjust VMError::segfault_address instead? It already adjusts for AIX, but apparently, that's not enough:
> 
> ```
> static constexpr intptr_t segfault_address = (1 * K) AIX_ONLY(+ (4 * K));
> ```
> 
> Note that this address is checked in jtreg (see test/hotspot/jtreg/runtime/ErrorHandling/TestSigInfoInHsErrFile.java), which you should adjust too (we expect the fault address to show up as fault address in the hs-err file).
> 
> This begs the question, if VMError::segfault_address is wrong today for AIX, TestSigInfoInHsErrFile.java must have failed, too, right?
> 
> Thanks, Thomas

Thanks @tstuefe for pointing this out. Updating the segfault_address is the better fix to this issue. The test/hotspot/jtreg/runtime/ErrorHandling/TestSigInfoInHsErrFile.java is also checking the segfault_address .I will update it.

-------------

PR: https://git.openjdk.org/jdk/pull/12468